Implications of Spatiotemporal Data Aggregation on Short-Term Traffic Prediction Using Machine Learning Algorithms
Short-term traffic prediction is a key component of Intelligent Transportation Systems. It uses historical data to construct models for reliably predicting traffic state at specific locations in road networks in the near future. Despite being a mature field, short-term traffic prediction still poses...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Wiley
2020-01-01
|
| Series: | Journal of Advanced Transportation |
| Online Access: | http://dx.doi.org/10.1155/2020/7057519 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832547115586289664 |
|---|---|
| author | Rivindu Weerasekera Mohan Sridharan Prakash Ranjitkar |
| author_facet | Rivindu Weerasekera Mohan Sridharan Prakash Ranjitkar |
| author_sort | Rivindu Weerasekera |
| collection | DOAJ |
| description | Short-term traffic prediction is a key component of Intelligent Transportation Systems. It uses historical data to construct models for reliably predicting traffic state at specific locations in road networks in the near future. Despite being a mature field, short-term traffic prediction still poses some open problems related to the choice of optimal data resolution, prediction of nonrecurring congestion, and the modelling of relevant spatiotemporal dependencies. As a step towards addressing these problems, this paper investigates the ability of Artificial Neural Networks, Random Forests, and Support Vector Regression algorithms to reliably model traffic flow at different data resolutions and respond to unexpected traffic incidents. We also explore different feature selection methods to identify and better understand the spatiotemporal attributes that most influence the reliability of these models. Experimental results indicate that data aggregation does not necessarily achieve good performance for multivariate spatiotemporal machine learning models. The models learned using high-resolution 30-second input data outperformed the corresponding baseline ARIMA models by 8%. Furthermore, feature selection based on Recursive Feature Elimination resulted in models that outperformed those based on linear correlation-based feature selection. |
| format | Article |
| id | doaj-art-d86270e92a3b45bea1e8e8598f46d3d1 |
| institution | Kabale University |
| issn | 0197-6729 2042-3195 |
| language | English |
| publishDate | 2020-01-01 |
| publisher | Wiley |
| record_format | Article |
| series | Journal of Advanced Transportation |
| spelling | doaj-art-d86270e92a3b45bea1e8e8598f46d3d12025-02-03T06:46:09ZengWileyJournal of Advanced Transportation0197-67292042-31952020-01-01202010.1155/2020/70575197057519Implications of Spatiotemporal Data Aggregation on Short-Term Traffic Prediction Using Machine Learning AlgorithmsRivindu Weerasekera0Mohan Sridharan1Prakash Ranjitkar2Department of Electrical and Computer Engineering, The University of Auckland, Auckland, New ZealandSchool of Computer Science, The University of Birmingham, Birmingham, UKDepartment of Civil and Environmental Engineering, The University of Auckland, Auckland, New ZealandShort-term traffic prediction is a key component of Intelligent Transportation Systems. It uses historical data to construct models for reliably predicting traffic state at specific locations in road networks in the near future. Despite being a mature field, short-term traffic prediction still poses some open problems related to the choice of optimal data resolution, prediction of nonrecurring congestion, and the modelling of relevant spatiotemporal dependencies. As a step towards addressing these problems, this paper investigates the ability of Artificial Neural Networks, Random Forests, and Support Vector Regression algorithms to reliably model traffic flow at different data resolutions and respond to unexpected traffic incidents. We also explore different feature selection methods to identify and better understand the spatiotemporal attributes that most influence the reliability of these models. Experimental results indicate that data aggregation does not necessarily achieve good performance for multivariate spatiotemporal machine learning models. The models learned using high-resolution 30-second input data outperformed the corresponding baseline ARIMA models by 8%. Furthermore, feature selection based on Recursive Feature Elimination resulted in models that outperformed those based on linear correlation-based feature selection.http://dx.doi.org/10.1155/2020/7057519 |
| spellingShingle | Rivindu Weerasekera Mohan Sridharan Prakash Ranjitkar Implications of Spatiotemporal Data Aggregation on Short-Term Traffic Prediction Using Machine Learning Algorithms Journal of Advanced Transportation |
| title | Implications of Spatiotemporal Data Aggregation on Short-Term Traffic Prediction Using Machine Learning Algorithms |
| title_full | Implications of Spatiotemporal Data Aggregation on Short-Term Traffic Prediction Using Machine Learning Algorithms |
| title_fullStr | Implications of Spatiotemporal Data Aggregation on Short-Term Traffic Prediction Using Machine Learning Algorithms |
| title_full_unstemmed | Implications of Spatiotemporal Data Aggregation on Short-Term Traffic Prediction Using Machine Learning Algorithms |
| title_short | Implications of Spatiotemporal Data Aggregation on Short-Term Traffic Prediction Using Machine Learning Algorithms |
| title_sort | implications of spatiotemporal data aggregation on short term traffic prediction using machine learning algorithms |
| url | http://dx.doi.org/10.1155/2020/7057519 |
| work_keys_str_mv | AT rivinduweerasekera implicationsofspatiotemporaldataaggregationonshorttermtrafficpredictionusingmachinelearningalgorithms AT mohansridharan implicationsofspatiotemporaldataaggregationonshorttermtrafficpredictionusingmachinelearningalgorithms AT prakashranjitkar implicationsofspatiotemporaldataaggregationonshorttermtrafficpredictionusingmachinelearningalgorithms |